Research on the English genitive (e.g. Rosenbach 2007: 154) reports increasing use of the s-variant. This has been explained as extension to inanimate possessors, a semantic shift (e.g. Hundt 1998; Rosenbach 2002), or due to the pressures of economy in journalism, a register change (Hinrichs & Szmrecsanyi 2007; Szmrecsanyi & Hinrichs 2008). The present work reports on a large-scale sociolinguistic investigation of the genitive in vernacular Canadian English using socially stratified corpora and individuals of all ages. The results show that human, prototypical possessors are 96 per cent s-genitive and non-humans are 95 per cent of-genitive. Within the small envelope where both forms are possible, we discover that variation patterns quite differently depending on animacy. For humans, use of the s-genitive is stable in apparent time and correlates with whether or not the possessor ends in a sibilant. In contrast, non-human collectives/organizations reveal an increasing use of s-genitives in apparent time and a favouring effect of short possessors, persistence (when an s-genitive has occurred recently in the previous discourse) and when the individual has a blue-collar job. Groups comprising humans (collectives and organizations), such as our church's youth group, and places that are possible locations for humans (countries, cities, etc.), as in Toronto's best restaurant, are the prime conduit for this change. These findings from vernacular speech confirm the extension of the s-genitive in inanimates by semantic extension.

This paper seeks to redefine apposition, a term that is often used in the literature with a remarkable lack of precision. Starting from paradigmatic instances of apposition (Romulus, the legendary founder; Santiago, the capital of Galicia), the main resemblances to the paradigm are analysed in an attempt to measure the validity of a general syntactic relation that is often put on a par with co-ordination and dependency. Paradigmatic appositions and other related nominal patterns are shown to be structures of nonrestrictive modification (i.e. not really appositions) which are best understood in reference to the concept of a Local Domain. The second unit of these structures has its scope in the local domain of the first unit, of which it is an expansion, not in the larger domain of the sentence. By contrast, most instances of non-nominal apposition (e.g. He ran – absolutely raced – up the hill; Burton-Roberts, 1975) are seen as true appositions, as defined in this paper, namely as structures whose two units relate independently to a Sentence Domain without forming a superordinate node in it. Crucial to the distinction between the Local and the Sentence domains is the role of intonation boundaries. These are strongly obligatory only with those structures which have been considered paradigmatic appositions in the literature (Romulus, the legendary founder), and their role is to isolate the second unit within them from the remainder of the sentence, thus preventing it from having a function in it. The intonation boundaries are also responsible for the most distinctive feature of these structures, namely the predicative relationship binding the two units together (Dupont, 1985; Koktová, 1985; Longrée, 1987; Forsgren, 1988).

This article explores measures, operationalisations and effects of rhythm and weight as two constraints on the variation between the s-genitive and the of-genitive. We base the analysis on interchangeable genitives in the news and letters sections of ARCHER (A Representative Corpus of Historical English Registers), which covers the period between 1650 and 1999. Thus, we are ultimately concerned with the applicability of two factors that have their roots in speech (rhythm: phonology; weight: online processing) to an ‘unconventional’, written data set with a historical dimension. As for weight, we focus on the comparison of simple single-constituent and more complex multi-constituent measurements. Our notion of rhythm centres on the ideally even distribution of stressed and unstressed syllables. We find that in our data set, both rhythm and weight show theoretically unexpected quadratic effects: rhythmically better-behaved s-genitives are not necessarily preferred over of-genitives, and short constituents exhibit odd weight effects. In conclusion, we argue that while rhythm is only a minor player in our data set, the quadratic quirks it exhibits should inspire further study. Weight, on the other hand, is a crucial factor which, however, likewise comes with measurement and modelling complications.

There are two important differences between American English and British English with respect to main verb have. First, American English typically employs do-support in constructions such as Do you have any coffee? while traditional British English does not. Secondly, American English typically does not use have in expressions such as I took a shower whereas British Isles English does: I had a shower. In this article, we discuss the possibility that there is a connection between these two facts. We argue that the connection lies in the failure of have in North American English to acquire the full range of dynamic meanings that it has acquired in other varieties of English, and suggest language contact as one explanation for this phenomenon.

This article concentrates on the question of language contact between English and Celtic in the period between the Anglo-Saxon conquest of Britannia (?AD 449) and the Norman conquest of England (AD 1066) but in some places reaches out to West Germanic times and to the period after the Norman conquest. It focuses on a certain region, that of the Southern Lowlands, mainly Anglo-Saxon Wessex, and deals with evidence that has been mentioned before: (1) the twofold paradigm of ‘to be’ and (2) the Old English designations for Celts that refer to their status as slaves. The article demonstrates that both the syntactic and the lexico-semantic evidence is particularly concentrated in West Saxon texts. Together, both types of evidence are shown to support the assumption that a very substantial Celtic population exerted substratal influence on (pre-)Old English by way of large-scale language shift in one of the early heartlands of England. This substratal Insular Celtic influence on Old English is contrasted with the adstratal Celtic influence on continental West Germanic.

Most handbooks and grammars contend that in Old English the voiced fricatives [v, ð, z] were merely allophones of /f, θ, s/ in sonorous environments. How these voiced fricatives became phonemes is debated among scholars. In this article, all previous accounts are critically reviewed. A new proposal is then presented, which explains the facts in a more direct way than previous theses. I argue that phonemicisation of a previous allophonic voice alternation in fricatives had already taken place in many areas of Anglo-Saxon England through language contact with Brittonic. Voiceless as well as voiced fricative phonemes existed in Brittonic at the time of contact, and language shift would have led directly to the phonemicisation of the previous allophonic variation found in early Old English.

This article examines the status of scalarity in the analysis of the meaning of the English determiner any. The latter's position as a prime exemplar of the category of polarity-sensitive items has led it to be generally assumed to have scalar meaning. Scalar effects are, however, absent from a number of common uses of this word. This suggests that any does not involve scales as part of its core meaning, but produces them as a derived interpretative property. The role of three factors in the derivation of the expressive effect of scalarity is explored: grammatical number, stress and the presence of gradable concepts in the NP. The general conclusions point to the importance of developing a causal semantic analysis in which the contributions of each of the various meaningful components of an utterance to the overall message expressed are carefully distinguished.

Following the Uniformitarian Principle, the Performance–Grammar Correspondence Hypothesis (PGCH; Hawkins 2004) predicts a directionality in language change: if the same content can be expressed by two competing structures and one of these is easier to process (see Hawkins 1999, 2004), then the simpler structure will be preferred in performance. Consequently, it will be used more often with a greater range of different lexical items, which increases its type frequency and ultimately leads to it being more cognitively entrenched than its alternative (see Hawkins 2004: 6). As an analysis of the diachronic evolution of the family of English comparative correlative constructions (the more iconic cause–before–effect C1C2 construction the more you eat, the fatter you get vs the less iconic effect–before–cause C2C1 construction you get the fatter, the more you eat) shows, however, the PGCH only played a secondary role in the genesis of this set of constructions. In this article, I will present a usage-based constructionist approach that allows researchers to reinterpret the classical Structuralist notion of gaps in the system as gaps in the mental constructional network. This type of Cognitive Structuralist analysis accounts for the presence of the less iconic C2C1 structure (and the absence of the more iconic C1C2 structure) in OE, the genesis of C1C2 structures at the end of the OE period as well as the processing effects predicted by the PGCH once both the C1C2 and the C2C1 constructions were in competition during the ME period.

This article surveys the debate on the origin of periphrastic do, with particular attention to the hypothesis that Celtic languages might have exerted some influence. With respect to the facts, it is argued that there are various types of Celtic hypotheses and that one type is sensible, though unlikely to be proven and even less likely to be proven to be the only relevant factor. With respect to the debate itself, it is shown that a Celtic origin hypothesis is accepted more widely among non-British scholars, and we speculate why that might be the case.

This article describes and discusses the appearance and increasing frequency of uh, um and er in American English journalistic prose from the 1960s to the early 2000s as part of the colloquialization of the language. The three variants uh, um and er are shown to have different uses in writing than in speech; in writing they can be shown to qualify as words, while their status in speech appears to be on a cline of wordhood. In writing, they belong to the class of stance adverbs, serving metalinguistic purposes. Two types are distinguished, depending on sentence placement: in initial position, uh, um and er are attitude adverbs and in medial position, they are style adverbs. Although er is dispreferred in initial position and preferred for correction of previously used words, every variant can be used for all discourse-pragmatic functions, which supports classifying them as one lexeme.

It is argued in this article that the postnominal ‘and adjective’ construction in Old English (e.g. Sodfßstne man & unscyldigne (LAW2,40.45) ‘righteous person and guiltless’ and pßas fullan ofnes & pßas peostran (BEDE,13.430.4) ‘that full furnace and that dark’) should be not analyzed as an instance of ambilateral adjective placement, but as an instance of DP coordination with an empty nominal element, pro, in the second conjunct. As a consequence, the adjective contained in the postnominal ‘and adjective’ construction is analyzed as an attributive adjective in prenominal, i.e. pre-pro position.

It has generally been assumed that Celtic linguistic influence on Old English is limited to a few marginal loanwords. If a language shift had taken place from Celtic to Old English, however, one would expect to find traces of that in Old English phonology and (morpho)syntax. In this article I argue that (1) the way in which the West Germanic sound system was reshaped in Old English strongly suggests the operation of a hitherto unrecognized substratum; (2) that phonetic substratum is strongly reminiscent of Irish rather than British Celtic; (3) the Old Irish phonetic−phonological system provides a more plausible model for reconstructing the phonetics of pre-Roman Celtic in Britain than the British Celtic system. The conclusion is that there is phonetic continuity between pre-Roman British Celtic and Old English, which suggests the presence of a pre-Anglo-Saxon population shifting to Old English.

This article examines the usefulness of Skousen's Analogical Modeling (AM) for explaining morphological change. In contrast to previous accounts of analogy, AM constitutes a general unified model of language that accounts for both sporadic and systematic changes. AM also provides explicit constraints on analogy that allow explanation of how morphological changes begin, which forms most likely serve as patterns for analogy, and which forms are most likely to change.

AM is then tested on the case of the adjectival negative prefix in English (in-, un-, dis-, etc.), using the Middle and Early Modern English portions of the Helsinki corpus as a basis for prediction. AM was given the task of using forms containing negative prefixes for one time period to predict the prefixes that adjectives would take in the subsequent time period. For each of the roughly seventy-year periods in the corpus, AM was able to predict valid prefixes about 90 percent of the time.

The nature and behaviour of complex and compound adverbs (e.g. very much, heretofore, anyway) has not received much scholarly attention in recent years. In the case of very much, for instance, recent literature (e.g. Dixon 2005) considers it a clause-internal adverb which typically modifies phrasal constituents (e.g. I liked the present very much; very much alive). The latter claim, however, appears to clash with previous observations (cf. Bolinger 1972) on the growing scope of the adverb in Present-day English. Through a corpus-based diachronic study (1500–present day), the present article unearths a number of environments where very much does not seem to fit neatly within the functional classifications that it has been assigned to in recent literature and standard grammars of English. It suggests that, from the Late Modern English period onwards (1800–), very much seems to have been developing sentence modifier functions, hence moving along Traugott's (1995) Internal Adverb > Sentence Adverb > Discourse Particle cline.

This article tries to fill a descriptive gap in the literature on English possessive NPs by offering a systematic description of indefinite possessive NPs (e.g. a friend's house, a widow's pension, men's suits). Specifically, two descriptive issues are dealt with. Firstly, the different types of determining indefinite genitives are discussed. It is shown that indefinite determiner genitives can have specific reference, generalized reference or generic reference. Moreover, it is argued that in all of these cases, the indefinite genitive specifies a reference point (Langacker 1993) for the identification of the ‘possessee’ and functions as a grounding predication. It is also proposed that subsective gradience (Aarts 2004, 2007) can be diagnosed for determining genitives, i.e. that more central and more peripheral members can be distinguished within this category. Secondly, the distinction between determining and nondetermining, classifying genitives (e.g. a [woman's magazine] ‘a magazine for women’) is re-examined. This distinction is relevant to the description of indefinite possessive NPs, since they have been considered as evidence for a fuzzy boundary between determining and nondetermining genitives (Taylor 1996, Rosenbach 2006). It is argued, against such an analysis, that the two functions of the genitive can be clearly distinguished, even in ‘bridging contexts’ (Evans & Wilkins 2000), i.e. contexts in which both a determining and a nondetermining reading are supported. Consequently, there is no intersective gradience (Aarts 2004, 2007) between determining and nondetermining genitives on the level of the constructional semantics.

The word such represents a challenge to most grammatical treatments when it comes to assigning it to a particular word class. In this article I consider the syntactic as well as the semantic properties of such. I argue that its conventional dual treatment as a determiner and as a pronoun (Biber, Johansson, Leech, Conrad & Finegan, 1999; Quirk, Greenbaum, Leech & Svartvik, 1985) is unwarranted. I also discuss some more recent treatments (Altenberg, 1994; de Mönnink, 1996; Siegel, 1994) and argue that these too cannot fully account for the facts. Based on syntactic evidence drawn from the British component of the International Corpus of English (ICE-GB), I will argue that the analysis I refer to as the ‘uniform adjective analysis’ of such (Huddleston & Pullum, 2002) best accounts for the facts.

This study examines sociophonetic variation in different functions of like among adolescents in London and Edinburgh. It attempts to determine the factors that may explain this variation. Our results suggest that the function of like correlates primarily with contextual factors, rather than the phonetic factors of vowel quality, /l/ to vowel duration and /k/ realisation. In particular, the preceding and following segments and their bigram predictability emerge as highly significant, in addition to the boundary strength following like. In both London and Edinburgh, the vowel appears to be the only non-contextual feature that is sensitive to the function of like: quotative be like is more likely to be monophthongised than other functions of like. We argue that the more monophthongal nature of quotative like is due to the syntactic and prosodic context in which it occurs.

There is a puzzling asymmetry in English with respect to free relative clauses introduced by what and who, with the former (e.g. [What Glenn said] didn't make much sense) intuitively being much more acceptable than the latter (e.g. [Who Glenn married] didn't make much money). In this squib, we explore this degraded acceptability of who free relative clauses, and from the results of an experimental study we identify syntactic features of the sentence that influence the level of acceptability. We discuss the difficulty in finding an independently motivated solution to the puzzling asymmetry within current theories of syntax, semantic and processing. Finally, we touch on a broader theoretical question relating to the robust cross-linguistic process by which elements of the set of wh-words in a language are able to extend their function from introducing interrogative clauses to introducing other clausal constructions.

Ellipsis alternation refers here to the alternation between two kinds of ellipsis remnants whose correlates are prepositional phrases. One kind of remnant includes the preposition hosted by its correlate and the other doesn't. This alternation is now known to be cross-linguistically widespread although it was originally assumed to be banned in languages without preposition stranding under wh-movement. I argue that there is a nonsyntactic relationship between ellipsis alternation and preposition stranding that helps explain the availability and distribution of both types of remnants in terms of general performance preferences. Two pieces of corpus evidence from English are offered in support of this argument. The first piece of evidence reveals that the content of a remnant and its correlate affects ellipsis alternation both in languages without preposition stranding and in English. The second piece of evidence shows that the availability of preposition stranding in English nonelliptical clauses supports the use of prepositionless remnants via structural persistence, that is, reuse of syntactic structure found in antecedent clauses. These data lead me to conclude that ellipsis alternation is subject to a stronger processing constraint in English than in languages without preposition stranding.

The Middle English period is well known as one of widespread lexical borrowing from French and Latin, and scholarly accounts traditionally assume that this influx of loanwords caused many native terms to shift in sense or to drop out of use entirely. The study analyses an extensive dataset, tracking patterns in lexical retention, replacement and semantic change, and comparing long-term outcomes for both native and non-native words. Our results challenge the conventional view of competition between existing terms and foreign incomers. They show that there were far fewer instances of relexification, and far more of synonymy, during the Middle English period than might have been expected. When retention rates for words first attested between 1100 and 1500 are compared, it is loanwords, not native terms, which are more likely to become obsolete at any point up to the nineteenth century. Furthermore, proportions of outcomes involving narrowing and broadening (often considered common outcomes following the arrival of a co-hyponym in a semantic space) were low in the Middle English period, regardless of language of origin.